🎬 Anime Recommendation System

📚 Conference Level Research

📋 Project Overview

Researching methods on giving recommendations for anime, based on its contents such as title and genres. The feature of the content will be extracted using BERT Method and compared using Cosine Similarity.

👨‍💻 My Role

In this project, I was responsible for conducting in-depth research on the model used, as well as writing and drafting our scientific paper. We developed an anime recommendation system that utilizes BERT to contextually understand anime descriptions, and then calculates similarity using Cosine Similarity.

🎤 Conference Presentation

I was the main presenter at the conference where we published the paper - representing the team to present the research and technical results of the model in front of academics and industry practitioners.

⚙️ Technical Implementation

BERT Model

Utilized BERT for contextual understanding of anime descriptions

Cosine Similarity

Implemented cosine similarity for accurate recommendation matching